Crystal Palace vs Arsenal

The model projects 9.07 corners. That sits well below the Premier League baseline of 10.5, and both team averages reinforce the suppressed read: Crystal Palace's home fixtures average 8.8 corners per game, Arsenal's away fixtures average 9.4. Two teams tracking beneath the league norm across recent samples is the central reason this assessment lands at medium confidence rather than high.

Variance is the complicating factor. Arsenal's away last-five reads 7, 13, 7, 7, 4, a sequence clustered at the low end with a single outlier at 13 that prevents the average from sitting where most of the data actually lives. Crystal Palace's home sequence, 15, 10, 6, 11, 10, is less extreme in its spread but still contains a 15 at the top end pulling the 8.8 average upward. When both sides carry dispersion of that kind in recent data, the model's projected total carries a wider band of uncertainty than a more stable input would produce.

The data is present. What it doesn't do is resolve cleanly: the spread between Crystal Palace's 8.8 and Arsenal's 9.4 is narrow, the predicted total sits beneath the league baseline of 10.5, and the variance in both last-five sequences prevents the kind of convergence that would sharpen confidence. Medium is the appropriate ceiling.

Stake discipline applies at this tier. Spectral's Premier League medium-confidence record: 101 settled, 46.5% win rate.